APPLICANT'S DESCRIPTION: Modification of regional brain glucose metabolism has been shown to occur with different classes of psychopharmacological compound. Different sites of localization and quantitative effects have been demonstrated using PET under controlled behavioral conditions. We propose to examine the sensitivity of this method by a within-class comparison of three antidepressant drugs in a placebo, double-blind, counterbalanced controlled study in human subjects. The basis of the study will be to generate statistically mapped brain images in common coordinate space, thus enabling neuroanatomic localization of the metabolic end effects of each drug to be identified. Previously acquired data demonstrating the metabolic changes produced by fluoxetine will be compared to venlafaxine and bupropion. We hypothesize that all three drugs will share a main effect in the limbic system, and in addition, venlafaxine and bupropion will show discernibly different metabolic end effects based on their specific mechanism of action. This approach represents a potentially significant contribution to drug development in clinical trials. In a cost-effective and efficient manner, it can provide predictive value for assessing brain effect and site effect profiles of new chemical entities and direct comparison with established within-class clinically effective drugs. PROPOSED COMMERCIAL APPLICATION: NOT AVAILABLE